Systems and methods for mapping neuronal circuitry and clinical applications thereof

Abstract

Systems and methods for mapping neuronal circuitry in accordance with embodiments of the invention are illustrated. One embodiment includes a method for generating a neuronal shape graph, including obtaining functional brain imaging data from an imaging device, where the functional brain imaging data includes a time-series of voxels describing neuronal activation over time in a patient's brain, lowering the dimensionality of the functional brain imaging data to a set of points, where each point represents the brain state at a particular time in the timeseries, binning the points into a plurality of bins, clustering the binned points, and generating a shape graph from the clustered points, where nodes in the shape graph represent a brain state and edges between the nodes represent transitions between brain states.

What is claimed is: 
     
       1. A method for virtually testing treatment using a neuronal shape graph, comprising:
 obtaining functional brain imaging data from an imaging device, where the functional brain imaging data comprises a time-series of voxels describing neuronal activation over time in a patient's brain; 
 lowering the dimensionality of the functional brain imaging data to a set of points, where each point represents the brain state at a particular time in the time-series; 
 binning the points into a plurality of bins; 
 clustering the binned points; and 
 generating a shape graph from the clustered points, where nodes in the shape graph represent a brain state and edges between the nodes represent transitions between brain states; 
 generate a virtual brain using the shape graph such that the virtual brain simulates the patient's brain; 
 virtually treating the virtual brain; 
 generating a second shape graph using the treated virtual brain; and 
 determining a treatment efficacy based on the second shape graph. 
 
     
     
       2. The method for generating a neuronal shape graph of  claim 1 , wherein lowering the dimensionality of the functional brain imaging data is performed using a neighborhood lens function. 
     
     
       3. The method for generating a neuronal shape graph of  claim 1 , wherein binning the points partitions the low dimensional points into overlapping bins. 
     
     
       4. The method for generating a neuronal shape graph of  claim 1 , wherein clustering the binned points is achieved using single-linkage clustering. 
     
     
       5. The method for generating a neuronal shape graph of  claim 1 , wherein the imaging device is a functional magnetic resonance imaging machine. 
     
     
       6. The method for generating a neuronal shape graph of  claim 1 , wherein the functional brain imaging data is resting state functional brain imaging data. 
     
     
       7. The method for generating a neuronal shape graph of  claim 6 , further comprising generating a Markov chain graph. 
     
     
       8. The method for generating a neuronal shape graph of  claim 1 , further comprising annotating the nodes in the shape graph with task data. 
     
     
       9. The method for generating a neuronal shape graph of  claim 1 , further comprising identifying at least one set of core nodes and at least one set of periphery nodes in the shape graph. 
     
     
       10. The method for generating a neuronal shape graph of  claim 1 , further comprising identifying a community structure within the shape graph. 
     
     
       11. A system for virtually testing treatment using a neuronal shape graph, comprising:
 at least one processor; and 
 at least one memory, comprising a neuronal mapping application, where the neuronal mapping application directs the processor to:
 obtain functional brain imaging data from an imaging device, where the functional brain imaging data comprises a time-series of voxels describing neuronal activation over time in a patient's brain; 
 lower the dimensionality of the functional brain imaging data to a set of points, where each point represents the brain state at a particular time in the time-series; 
 bin the points into a plurality of bins; 
 cluster the binned points; and 
 generate a shape graph from the clustered points, where nodes in the shape graph represent a brain state and edges between the nodes represent transitions between brain states; 
 generate a virtual brain using the shape graph such that the virtual brain simulates the patient's brain; 
 virtually treat the virtual brain; 
 generate a second shape graph using the treated virtual brain; and 
 determine a treatment efficacy based on the second shape graph. 
 
 
     
     
       12. The system for generating a neuronal shape graph of  claim 11 , wherein the neuronal mapping application further directs the processor to lower the dimensionality of the functional brain imaging data is performed using a neighborhood lens function. 
     
     
       13. The method for generating a neuronal shape graph of  claim 1 , wherein the bins are overlapping. 
     
     
       14. The method for generating a neuronal shape graph of  claim 1 , wherein the neuronal mapping application directs the processor to cluster the binned points using single-linkage clustering. 
     
     
       15. The method for generating a neuronal shape graph of  claim 1 , wherein the imaging device is a functional magnetic resonance imaging machine. 
     
     
       16. The method for generating a neuronal shape graph of  claim 1 , wherein the functional brain imaging data is resting state functional brain imaging data. 
     
     
       17. The method for generating a neuronal shape graph of  claim 6 , wherein the neuronal mapping application further directs the processor to generate a Markov chain graph. 
     
     
       18. The method for generating a neuronal shape graph of  claim 1 , wherein the neuronal mapping application further directs the processor to annotate the nodes in the shape graph with task data. 
     
     
       19. The method for generating a neuronal shape graph of  claim 1 , wherein the neuronal mapping application further directs the processor to identify at least one set of core nodes and at least one set of periphery nodes in the shape graph. 
     
     
       20. The method for generating a neuronal shape graph of  claim 1 , wherein the neuronal mapping application further directs the processor to identify a community structure within the shape graph.